Predicting Suicide: 2% of Depressed and 1.6% of Population Die by Suicide

The big question that often arises for many of us is what can we do to prevent suicide? Sadly, not much, according to experts.  We've known that much since at least 2016 when researchers studied risk factors depression, previous suicide attempts, stressful life events and substance abuse and expert's ability to predict suicidal thoughts and behaviors over long periods of time.

Predicting if someone will attempt to take his or her own life is only slightly better than chance and has not significantly improved during the last five decades.  Those findings still hold true, said Joseph Franklin, an assistant professor of psychology at Florida State University and lead author of the study.

Franklin and his colleagues conducted a meta-analysis of 365 studies conducted within the last 50 years.  The analyses, he said, showed a suicide expert who conducted an in-depth assessment of risk factors would predict a patient's future suicidal thoughts and behaviors with approximately the same degree of accuracy as someone with no knowledge of the patient.  "It's basically on par with flipping a coin," he said. "It turns out, for example, things like depression don't predict much better than random guessing either."

Franklin said, for instance, just 2 percent of people who are severely depressed will die by suicide. But 1.6 percent of the general population around the world and in the U.S. will eventually die by suicide.  "There are thousands of things relevant to suicide, but there's no one thing among those thousands that stand out as being particularly important for prediction," he said.

Still, Franklin said the findings do not necessarily mean that widely used risk guidelines are invalid or useless, or that therapists should abandon them.  "We recommend that these guidelines remain in use, but emphasize that there is an urgent need to evaluate these guidelines within longitudinal studies," he said.  Nadine J. Kaslow, a tenured professor in the Emory University School of Medicine and past president of the American Psychological Association, agreed traditional risk factors such as depression, substance abuse, stress or previous suicide attempts aren't adequate predictors of suicide by themselves.

What's happening in a person's life at the moment a job loss, relationship break up, or mental health or substance abuse problems combined with some current stressor might be, she said. At a time when people are desperately searching for ways to tell if a loved one is in trouble, Kaslow said it's important that people not blame themselves if they weren't able to prevent someone from dying by suicide.  "To convey that there is nothing that can be done, will scare people even more and make them feel more helpless," Kaslow said. "We're not saying that people do nothing. It doesn't mean we don't keep trying to figure out what to do. If someone is struggling, we need to reach out to them. We do know that strong social support protects people from attempting suicide or dying by suicide."

Franklin is hopeful a new approach to predicting suicide is in our future.  For instance, he said, multiple groups have begun working on developing "machine learning algorithms" the same things that drive the Google Search algorithm, make your email spam filter effective and show you relevant advertisements to combine tens or even hundreds of risk factors together to predict suicidal behaviors.  "The preliminary results are promising, with algorithms predicting suicidal behaviors with 80-90 percent accuracy," Franklin said. "The big catch though is translating that to clinical practice will take a few years, and based on current studies, we can tell who will eventually die by suicide but not when."

Meanwhile, he said, the only things that consistently reduce risk are large-scale preventative measures, such as reducing the availability of guns, putting pills in blister packs, or erecting fences around certain bridges.